高分辨率遥感影像快速去雾

doi:10.11947/j.AGCS.2022.20200480

摘要/Abstract

摘要： 针对含雾遥感影像在军事航空侦查、地物判读等方面使用率低、有效性差的问题，以及现有去雾算法中存在计算复杂耗时、色彩失真的弊端，结合遥感影像景深变化小、不含天空背景等特点，本文提出一种改进的暗原色先验去雾算法。首先，对影像中白色场景灰度值进行统计并设定阈值划分为失效区，分离水域与非水域减少蓝色波段在水域的占比，合成新的蓝色波段，以改进暗通道值的获取方法；其次，采用导向滤波替代软抠图法优化透射率提升处理时间；然后，对关键参数进行适应性改进试验并采用自动色阶恢复去雾后的影像色彩；最后，利用含雾的无人机影像和GF-2影像进行了试验，并进行了定量评价。试验结果表明，在同等试验条件下，本文方法处理单幅影像的时间比暗原色先验去雾算法的提升4倍以上，且去雾后影像的灰度均值、标准差、信息熵、平均梯度等指标比暗原色先验去雾算法得到的值均有提高，能有效提高有雾影像的清晰度，增强影像色彩和细节。

关键词: 遥感影像, 去雾, 暗原色先验, 导向滤波

Abstract: Aiming at the problems of low utilization rate and poor effectiveness of cloud and fog remote sensing images in military aviation investigation and ground object interpretation, as well as the disadvantages of the existing cloud and fog removal algorithms, such as complex calculation, time-consuming and color distortion, combined with the characteristics of remote sensing image with small depth of field change and without sky background, an improved dark-channel prior dehazing algorithm is proposed. Firstly, the gray value of the white scene in the image is counted and the threshold is set to divide it into failure area. The water area and non water area are separated to reduce the proportion of blue band in the water area, and a new blue band is synthesized to improve the acquisition method of dark channel value; Secondly, the guided filter is used to replace the soft matting method to optimize the transmittance enhancement processing time; then, the adaptive improvement experiment of key parameters is carried out and the automatic color level restoration is adopted the color of the image after defogging. Experiments are carried out with fog UAV images and GF-2 images, and quantitative evaluation is carried out. The experimental results show that under the same experimental conditions, the processing time of a single image by this method is more than 4 times higher than that of the dark-channel prior algorithm, and the gray mean, standard deviation, information entropy and average gradient of the dehaze image are higher than those obtained by the dark-channel prior algorithm, which can effectively improve the clarity of the fogged image, Enhance image color and detail.

Key words: remote sensing image, dehaze, dark-channel prior, guide filter

中图分类号:

P237

廖章回, 姜闯. 高分辨率遥感影像快速去雾[J]. 测绘学报, 2022, 51(3): 446-456.

LIAO Zhanghui, JIANG Chuang. Fast dehaze of high resolution remote sensing images[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(3): 446-456.

参考文献

[1] 梁栋,孔颉,胡根生,等. 基于支持向量机的遥感影像厚云及云阴影去除[J].测绘学报,2012,41(2):225-231, 238. LIANG Dong,KONG Jie,HU Gensheng,et al. The removal of thick cloud and cloud shadow of remote sensing image based on support vector machine[J]. Acta Geodaetica et Cartographica Sinica,2012,41(2):225-231, 238.
[2] 陈振炜,张过,宁津生,等. 资源三号测绘卫星自动云检测[J].测绘学报,2015,44(3):292-300.DOI:10.11947/j.AGCS.2015.20130384. CHEN Zhenwei,ZHANG Guo,NING Jinsheng,et al. An automatic cloud detection method for ZY-3 satellite[J]. Acta Geodaetica et Cartographica Sinica,2015,44(3):292-300.DOI:10.11947/j.AGCS.2015.20130384.
[3] DAI Yuchao, ZHANG Jing, HE Mingyi, et al. Salient object detection from multi-spectral remote sensing images with deep residual network[J]. Journal of Geodesy and Geoinformation Science,2019,2(2):101-110.
[4] 吴迪,朱青松.图像去雾的最新研究进展[J].自动化学报,2015,41(2):221-239. WU Di,ZHU Qingsong. The latest research progress of image dehazing[J]. Acta Automatica Sinica, 2015,41(2):221-239.
[5] MCCANN J J.Retinex at 40[J]. Journal of Electronic Imaging,2004,23(1):6-14.
[6] 郭瑞,党建武,沈瑜,等. 改进的单尺度Retinex图像去雾算法[J]. 兰州交通大学学报,2018,37(6):69-75. GUO Rui, DANG Jianwu,SHEN Yu,et al. Fog removal algorithm of improved single-scale retinex image[J]. Journal of Lanzhou Jiaotong University,2018,37(6):69-75.
[7] SHEN Huanfeng, LI Huifang, QIAN Yan, et al. An effective thin cloud removal procedure for visible remote sensing images[J]. ISPRS Journal of Photogrammetry & Remote Sensing,2014,96(11):224-235.
[8] 朱锡芳,吴峰,庄燕滨. 小波变换在遥感图像云雾处理中的应用[J]. 微电子学与计算机,2006,23(12):50-52. ZHU Xifang,WU Feng,ZHUANG Yanbin.Thin cloud and mist reduction from remote sensing images based on wavelet transformation[J]. Microelectronics and computer,2006,23(12):50-52.
[9] 朱锡芳,吴峰,庄燕滨. 基于Mallat算法遥感图像去云雾处理的改进方法[J]. 遥感学报,2007,11(2):241-246. ZHU Xifang,WU Feng,ZHUANG Yanbin. An improved approach to remove cloud and mist from remote sensing digital images based on mallat algorithm[J]. Journal of Remote Sensing,2007,11(2):241-246.
[10] 陈奋,闫冬梅,赵忠明. 基于无抽样小波的遥感影像薄云检测与去除[J]. 武汉大学学报(信息科学版),2007,32(1):71-74. CHEN Fen, YAN Dongmei, ZHAO Zhongming. Haze detection and removal in remote sensing images based on undecimated wavelet transform[J]. Geomatics and Information Science of Wuhan University, 2007, 32(1):71-74.
[11] SINGH D, KUMAR V. A novel dehazing model for remote sensing images[J]. Computers and Electrical Engineering, 2018, 69(5):14-27.
[12] LIU Qi, GAO Xinbo, HE Lihuo, et al. Haze removal for a single visible remote sensing image[J]. Signal Processing,2017,137:33-43.
[13] 陆婉芸,王继周,曹萌.变序字典学习AO-DL的资源三号遥感影像云去除[J]. 测绘学报,2017,46(5):623-630. DOI:10.11947/j.AGCS.2017.20160474. LU Wanyun,WANG Jizhou,CAO Meng. Cloud removal in ZY-3 remote sensing image based on atoms-reordered dictionary learning AO-DL[J]. Acta Geodaetica et Cartographica Sinica,2017,46(5):623-630.DOI:10.11947/j.AGCS.2017.20160474.
[14] HE Kaiming, SUN Jian, TANG Xiaoou. Single image haze removal using dark channel prior[C]//Proceedings of the Computer Society Conferenceon Computer Visionand Pattern Recognition. Miami,USA:IEEE,2009:1956-1963.
[15] LONG Jiao, SHI Zhenwei, TANG Wei. Fast haze removal for a single remote sensing image using dark channel prior[C]//Proceedings of International Conference on Computer Vision in Remote Sensing. Xiamen,China:IEEE, 2012.
[16] 王时震,万惠琼,曾令沙,等. 应用暗原色先验规律的遥感影像去雾技术[J]. 测绘科学技术学报,2011,28(3):182-185. WANG Shizhen,WAN Huiqiong,ZENG Lingsha,et al. Remote sensing image defogging technology based on the prior law of dark primary color[J]. Journal of Surveying and Mapping Science and Technology,2011,28(3):182-185.
[17] 汤群芳,杨杰,刘海波, 等. 基于暗通道先验的单幅图像快速去雾方法[J]. 光子学报,2017,46(9):205-213. TANG Qunfang,YANG Jie,LIU Haibo,et al. Fast single-image dehazing method based on dark channel prior[J]. Acta Photonica Sinica,2017,46(9):205-213.
[18] YUAN Quan, SHEN Huanfeng, LI Huifang. Single remote sensing image haze removal based on spatial and spectral self-adaptive model[J]. Image and Graphics,2015,92(19):382-392.
[19] 赵桂华,邹晓亮,戴晨光,等. 基于暗原色的航空遥感影像薄雾去除与改进[J]. 测绘科学,2013,38(1):29-32. ZHAO Guihua,ZOU Xiaoliang,DAI Chenguang,et al. An improved method of haze removal for aerial digital image based on dark channel prior[J]. Science of Surveying and Mapping,2013,38(1):29-32.
[20] 杨靖宇,张永生,邹晓亮,等.利用暗原色先验知识实现航空影像快速去雾[J].武汉大学学报(信息科学版),2010,35(11):1292-1295. YANG Jingyu,ZHANG Yongsheng,ZOU Xiaoliang,et al.Using dark channel prior to quickly remove haze from a single image[J]. Geomatics and Information Science of Wuhan University,2010,35(11):1292-1295.
[21] 甘玉泉,汶德胜,王乐,等.暗通道自然灾害遥感图像去雾[J].光子学报,2015,44(6):53-57. GAN Yuquan,WEN Desheng,WANG Le,et al. Haze removal using dark channel for remote sensing images of natural disaster[J]. Acta Photonica Sinica,2015,44(6):53-57.
[22] 王京卫. 测绘无人机低空数字航摄影像去雾霾研究[J]. 测绘学报,2016,45(2):251.DOI:10.11947/j.AGCS.2016.20150412. WANG Jingwei. Study of the Geo-UAV low-altitude digital aerial image haze removal[J].Acta Geodaetica et Cartographica Sinica,2016,45(2):251.DOI:10.11947/j.AGCS.2016.20150412.
[23] 陆士猛,刘昌锦. 无人机侦察图像快速去雾算法[J].红外技术,2015,37(10):847-851. LU Shimeng, LIU Changjin. The fast haze removal algorithm for UAV reconnaissance image[J]. Infrared Technology,2015,37(10):847-851.
[24] NARASIMHAN S, NAYAR S. Vision and the atmosphere[J]. International Journal of Computer Vision,2002,48(3):233-254.
[25] MCFEETERS S K. The use of the normalized difference water index (NDWI) in the delineation of open water features[J]. International Journal of Remote Sensing,1996,17(7):1425-1432.
[26] HE Kaiming, SUN Jian, TANG Xiaoou.Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1397-1409.